Is political science a science? The answer to this question demands far more nuance than a simple yes or no could ever suffice to give. Instead, what is needed is a fundamental definition of what science is in order to understand the compatibility that does or does not exist between the two. In its most rudimentary form, we may define science as the systematic method by which we accumulate knowledge, with knowledge further being defined as justified true belief. Guided by the form of this definition, I intend to argue that yes, political science can very well systematically accumulate knowledge and (modus ponen) be considered a science. But the form of this definition and the conclusion I bring with it demands an epistemic clarification upon how exactly political science can do this.
The hard sciences, ignited by the catalyst of the radical empiricism of Francis Bacon, Isaac Newton, and John Locke, first began making prodigious successes through what was thought to be predominantly systematic observations and inductive reasoning. This success, argued the new positivists of the 20th century could be imported to the social sciences, doing away with the metaphysicians and social philosophers and replacing them with scientist’s who make empirically falsifiable observations (Bernstein 1978, 5). The influence of the positivist movement in political science cannot be understated, impressively dominating the current literature with heavily quantitative research. But is the positivistic imitation of the hard sciences appropriate for political science? Hardly not. Political Science is unique to the hard sciences of physics, chemistry, biology, geology, etc. not in a fundamental way, but none the less in quite drastic form.
To further apprehend why, we must better understand the inductive method of empirical knowledge and its inescapable Humean problem of inscrutable causality. Inductive reasoning is the process by which specific instances are utilized to develop general rules. Dropping a ball and watching it move towards the earth at a steady rate, followed by concluding a general theory of gravity is a form of inductive reasoning. No matter how many times we experience specific cases of something, it does not logically preclude something different from happening the next time the specific case is observed. Of course, all Hume’s argument showed was that we could not logically conclude with certainty the general rule from any specific observation, but this is not to say the empirical method of induction should be abandoned; in the hard sciences such as physics, the inductive method brings further certainty each time a correlation is experienced or not experienced. But when we attempt to use the inductive method empirically towards the social sciences, as some political scientists like Robert Lane would have us to do (Lane 763), problems exacerbate.
The human mind has over 86 billion neurons with the capacity to compute about 17.2 trillion calculations per second (Houzel-Herculano), making it by far the most complicated and difficult thing in the known universe. Beyond the sheer complexity of the brain itself and the internally caused complex behaviors its produces, human beings also actively engage with an incalculable amount of other variables such as the individual social relations of family members, friends, co-workers, and business partners; the larger social and cultural pressures at national, regional, and local levels; physical influences such as sickness, weather, technology, and natural disasters; and a whole slew of other variables, some of which may yet still be unknown. Unlike the hard sciences that deal with the laws of nature, which are relatively static and have variables that are relatively easy to demarcate, the social sciences deal with human beings who are far less static and possess variables which are far less easily demarcated. Robert Merton, largely taking from Kant, argues that “Empirical research without theory is blind, just as theory without empirical research is empty” (Bernstein 1978, 14). Collecting data and proposing ad hoc hypotheses based off of correlations will not be sufficient in the social sciences because the large amount of variables involved leaves little hope of ever coming across true knowledge in such a blind fashion.
Since induction is blind and rather helpless to the social scientists, what is to be done with the empirical knowledge given to us as political scientists without the correct theoretical categories to organize it? Robert Merton and other empirical theorists think a “hypothetical-deductive” model is the answer, combining the far more suitable deductive method with empirical observations to accumulate knowledge (Bernstein, 13). But where does one come up with the premises that make up a hypothetical deduction? The answer is through empirical induction, which means we have not solved the positivists problem, but only pushed the issue back one step.
As was so famously brought to light in the metaphysical work of Immanuel Kant and his Copernican revolution, and also well elaborated by Alfred Schutz, we have discovered that man does not passively take in the world around him but instead imposes structure and rationality to it (Bernstein, 147). Similar to the Kuhnian paradigms which allow for more and more “puzzles” to be solved by actively adopting and discarding different theoretical frameworks (Kuhn, 175), what is needed in political science is a theory that we can impose upon the field to bring understanding and sensibility to the social world. But this theory cannot rely upon the ad hoc inductive processes mentioned above, since innumerable variables and correlates would never provide the type of certainty needed for justified true belief.
Instead, the answer to our epistemic problems rest within the logical structure of the human mind, available to us a priori. As is largely argued by phenomenologists like Schutz, when we evaluate our own minds, we see certain logical structures that exist as permanent ahistorical categories which can be used to organize the empirical world around us (Bernstein, 147). For example, we can discover and confirm the validity of the human action axiom, which states that all human beings purposefully act by choosing means to achieve their most desired ends, a priori. This axioms validity is known a priori because the very act of denying the axiom involves utilizing the means of argumentation to achieve the ends of denying the axiom which produces an internal contradiction (Mises 2008). Many of the tools of the rational choice theorists such as methodological individualism and the theory of purposeful action fall into these logical categories as a necessary part of our being (Ordeshook, 1). Using the theoretical tools learned a priori about the logical structure of the human mind as premises, the method of deduction or a far more helpful induction can be applied to the social sciences similar to that done by the rational choice theorists.
But we must stipulate caution under such an epistemological understanding, for the breadth and extent of the logical categories of the human mind can only take us so far. The rational choice assumptions of fixed preferences (or ends), cardinal utilities, and narrow materialistic utility maximization, while crucial to producing highly predictive models, do not contort with reality. Our minds value things in the ordinal, which changes over time, dependent upon the flux of personal and social variables one deals with at any point in time. These changing values also relate to the ends we choose which in no way have to be narrowly and materialistically self-interested but may be much more broad and other regarding (Ostroom, 17).
Among the many variables that influence the manner in which we impute value onto our preferences and interpret the world is what James Farr spoke of as the historical ones. It is true that conceptions such as “revolution”, and the causal influence such conceptions have upon us, have different meanings to different cultures and different historical time-periods throughout human history (Farr, 698). But unlike the rational choice theorists, who make erroneous assumptions about human behavior for predictive purposes, Farr creates an historical chasm of flux which is impenetrable and devoid of giving theoretical explanatory power to historical events or predictive power to future events. This method is just as erroneous as it ignores the logical structure that the human mind provides as an objective and static link between past, present, and future. By taking the empirical knowledge of what was meant by “revolution” at any given time, we can use the theory of the logical structure of the mind, an ahistorical framework, to make historical events intelligible and relatable within present contexts. For example, once we understand the theory of time preference, or that people desire things in the present over the same things in the future, as a logical a priori category of our human nature, we can help structure and explain historical events in the past by means of this tool and connect them with how we currently act in the present.
Our understanding of the structure we bring to the world around us can also assist other epistemic approaches to political science, like critical theory, by providing the substructure upon which a critical theorist can stand and objectively evaluate the socio-political influences that causally influence people and social institutions. The Marxist’s of the Frankfurt school could only argue that social classes influence the way in which man understands himself if marxists themselves have an objective theoretical tool to apply across different classes comparatively (Galtung, 19). Otherwise, they could not escape the indictment that their theory was the result of class consciousness and has no validity outside of that very class. More importantly for the critical theorists, similar to Kant’s deontological ethics, understanding the a priori logic of our nature provides ethical norms that can be judged objectively by the law of non-contradiction and applied universally without fear of an empirical subjectivism of some sort.
Can political science accumulate knowledge? I hope to have answered in the affirmative and to have demonstrated the unique way in which the social sciences can do this. Unlike the hard sciences of physics, chemistry, etc. we have cognitive access to our own minds and certain insights into why we do what we do. Although this access is limited, for many of our values are socially and subconsciously constructed for us behind the scenes of our cognition, there are important insights to be made. And thankfully so, for the empirical method so successful in the physical science, just cannot be so in the social sciences with the vast flux of variables that must be dealt with. Instead, we have the capacity of knowing, a priori, the structures of the human mind to impose order upon all the variables of the social world to accumulate knowledge to its furthest possible degree.
Works Cited
Bernstein, Richard. 1976. The Restructuring of Social and Political Theory. US: University of Pennsylvania Press.
Farr, James. 1982. “Historical Concepts in Political Science: The Case of ‘Revolution’” American Journal of Political Science Vol. 26, No. 4 (Nov., 1982), pp. 688-708. Galtung Johan. 1977. “Positivism and Dialectics: A Comparison.” Essays in Methodology. Copenhagen: Ejlers Houzel-Herculano, Suzana. Laboratory of Comparative Neuroanatomy. http://www.suzanaherculanohouzel.com/lab. University of Federal do Rio de Janeiro. 2009. Web. Accessed 5/15.
Kuhn, Thomas. The Structure of Scientific Revolutions. 2nd ed. Chicago: University of Chicago Press.
Lane, Robert. 2003. “Rescuing Political Science from Itself.” In David O. Sears, Leonie Huddy, and Robert Jervis eds. Oxford Handbook of Political Psychology. Oxford: Oxford University Press. Ch. 21, pp. 755-793.
Mises, Ludwig. 2008. Human Action. US: Ludwig Von Mises Institute.
Ordeshook, Peter. 1986. Game Theory and Political Theory. Cambridge: Cambridge University Press.
Ostrom, Elinor. 1997. “A Behavioral Approach to the Rational Choice Theory of Collective Action.” The American Political Science Review, Vol. 92, No. 1 (Mar 1998), pp. 1-22.